Suicide is a leading cause of death and suicide attempts are a major cause of disability, lost productivity, and health care costs. Suicide prevention is a research priority of the National Institutes of Health, and the US Surgeon General's National Strategy for Suicide Prevention calls for a shift towards recovery-oriented prevention efforts which promote hope and social support. Hopelessness and social isolation are two proximal risk factors for suicide which may be improved via peer mentorship, a form of peer support effective for preventing depression and repeat psychiatric hospitalizations. The primary aims of this study are to develop and pilot test a peer mentorship intervention for psychiatrically hospitalized patients at high risk for suicide. The intervention will be adapted by an expert panel from existing peer support training protocols to target suicide risk factors and to enhance suicide risk management. Protocols for training and supervising peer mentors and measures of intervention fidelity will also be developed. The intervention will then be pilot teste among 60 participants randomly assigned to receive the peer mentorship intervention plus usual care or usual care alone. Participants will be recruited from the inpatient psychiatry unit at the University of Michigan Health System. Inclusion criteria will include medical record documentation of suicidal ideation or suicide attempt at admission, and exclusion criteria will include significant cognitive impairment (according to the Mini-Cog), current receipt of peer support, or determination that peer mentorship may cause distress to the patient or the peer mentor. The peer mentorship intervention will include an in-person visit on the inpatient unit and regular in-person or telephone follow-up for 3 months post-discharge. The intervention will be delivered by peer specialists--individuals in stable recovery from serious mental illness who have received formal training and certification in peer support from the state of Michigan--with at least 6 months of professional peer support experience. The primary outcomes of the pilot study are acceptability and feasibility of the intervention as determined by: 1) >50% of eligible participants enroll in the study, 2) >70% of enrollees complete final follow- up measures at 6 months, and 3) among those assigned to the peer mentorship intervention, >80% complete an inpatient session and the median number of total sessions is at least 4. Peer mentorship sessions will be recorded and rated for fidelity. Measures of suicidal ideation and suicide attempts (the intended primary outcomes of a subsequent efficacy study) and secondary outcomes such as quality of life, functioning, depression, and service use will be obtained at baseline, 3 months, and 6 months post-enrollment by a research assistant blinded to study arm. An exploratory aim will be to measure potential mediators of intervention effectiveness including belongingness, burdensomeness, and hopelessness according to the interpersonal theory of suicide. If acceptability and feasibility are demonstrated, the study will result in a novel recovey-oriented suicide prevention intervention ready for a fully-powered randomized controlled efficacy trial.